节点文献
基于光谱和径向基函数神经网络的空间碎片识别
Space debris Recognition based on spectra and RBF neural network
【摘要】 建立了以光谱望远镜和图像跟踪系统相结合的空间碎片测量装置,对空间碎片进行光谱测量。对获得的1 007个样本数据进行离散二进小波变换和矢量归一化处理分析,得到了噪声低、相对强度分布集中的光谱数据。用70%光谱数据进行训练得到径向基函数神经网络模型,以剩余的30%数据测试模型的准确度,模型准确率达到88.74%。
【Abstract】 A space debris detector is established with spectroscopic telescope and image tracking system to measure the spectrum of space debris.Discrete dyadic wavelet transform and vector normalization are applied to analyze the 1007 samples,and a set of data with low noise and concentrated relative intensity distribution are obtained.70% of the data are trained for getting the radial basis function neural network model while the other 30% for testing the model.The results show that the accuracy rate of the model is 88.74%.
【基金】 吉林省科技发展计划基金资助项目(20130101179JC,20140101199JC,20150101053JC)
- 【文献出处】 长春工业大学学报 ,Journal of Changchun University of Technology , 编辑部邮箱 ,2015年05期
- 【分类号】TP183;TP391.4
- 【下载频次】111